Lifetime prevalence of anxiety disorders in people with bipolar disorder: A systematic review and meta-analysis

Background: Anxiety disorders are increasingly recognised as an important determinant of outcomes in patients with bipolar disorder. However, a reliable estimate of their prevalence is still missing, because the published prevalence of anxiety disorders in individuals with bipolar disorder varies widely. In this study, we aimed to quantify the lifetime prevalence of anxiety disorders in individuals with bipolar disorder and compare it with rates in people without the disorder. Methods: We searched the Web of Knowledge and Medline (through the PubMed interface) for articles published in any language from the database inception dates up until June 1, 2014, using a combination of the word "bipolar" and search terms for anxiety disorders. We included studies that reported original data about the lifetime prevalence of DSM-III and DSM-IV anxiety disorders in adults with bipolar disorder that recruited participants irrespective of comorbidities and that used a validated diagnostic interview to establish the diagnoses of bipolar disorder and at least one anxiety disorder. We excluded studies that reported only the current prevalence or if we were unable to establish whether they described current or lifetime prevalence, and those with discrepancies in the data that could not be resolved by contacting the authors. We did a random-effects meta-analysis of lifetime prevalence of DSM-III and DSM-IV anxiety disorders in adults with bipolar disorder, in which we quantified the lifetime prevalence of any anxiety disorder in people with bipolar disorder. We compared this prevalence in people with bipolar I disorder versus those with bipolar II disorder, and in people with bipolar disorder versus population controls. Findings: Data from 40 studies, including 14 914 individuals from North America, Europe, Australia, South America, and Asia, indicate that the lifetime prevalence of anxiety disorders in individuals with bipolar disorder is 45% (95% CI 40-51). Direct comparison in five samples with a total of 1378 individuals with bipolar disorder and 56 812 population controls without bipolar disorder indicates a three-fold increase (risk ratio [RR] 3·22 [95% CI 2·41-4·29]; p<0·0001) in the prevalence of anxiety disorders in those with bipolar disorder. 13 studies that included both individuals with bipolar I disorder (n=4270) and those with bipolar II disorder (n=1939) showed no difference in the lifetime prevalence of anxiety disorders between these subtypes (RR 1·07 [95% CI 0·96-1·20]; p=0·223). We noted significant heterogeneity among included studies that was not accounted for by reported differences in study characteristics. Interpretation: People with bipolar disorder are at increased risk of anxiety disorders compared with those without bipolar disorder; nearly one in two has an anxiety disorder in their lifetime. Anxiety disorders should therefore be assessed alongside the mood symptoms in patients with bipolar disorder. Funding: Capital Health Research Fund.